Earlier this week, Microsoft introduced new AI additions to its suite of software with Microsoft 365 Copilot, a large language model used to cut down on the daily slog of emails and other menial tasks. Now, this tooling is being extended to Microsoft Power Apps, to help people actually create or augment apps by simply describing what they need in natural language.
Available for several years, Microsoft Power Apps is a way to “rapidly build and share low-code apps” without having to invest in complex development or upkeep to a significant degree. Between templates, and the ability to drag-and-drop content and tweak as needed, Power Apps enables users to go further with problems and streamline processes that could use automation. Now, though, Microsoft is kicking things up a notch with the use of AI copilot in Power Apps.
Using the recent advancements in AI and large language models, app makers can “build an app, including the data behind it, just by describing what you need through multiple steps of conversation.” While this isn’t necessarily brand new, first being shown in 2021 with suggestions rather than full app creation, it is still interestinig and very cool. For example, creators can say things like, “Build an application to automate the manual process of creating and approving customer invoices,” and Copilot will take it from there.
With a little back and forth with Copilot, you can end up with an app customized to your needs. You can also use Copilot to fill in sample data or get suggestions about what might be good for the app in the future. Moreover, AI assistance can help query data as well, finding common holes or insights into data based on questions like “What are the most common reasons an inspection fails?”
It seems the days of simple apps and tools are behind us as options like Power Apps take center stage. Between this and ChatGPT passing exams and explaining memes, humans will be better able to focus on more complex problems that AI cannot yet solve. However, it might not be much longer before we don’t have to do much in terms of hand coding, and that could be a good thing.